This proposal is designed to generate preliminary data useful for future studies examining the natural history of the process of disability and will serve as a preclinical screening tool for future pharmacological interventions. The rationale for this proposal comes from studies in humans showing that declining performance on a range of physical performance measures later predicts the incidence of disability in activities of daily living (or ADLs) and mortality. These declines are associated with naturally occurring, agerelated physiological changes in body composition and strength known as sarcopenia. Early detection and characterization of this condition and its relationship to declines in physical performance will aid in the development of pharmacological interventions. Therefore, the specific aim of this proposal is to establish, in animals, the predictive validity of a procedure for assessing physical performance and disability measures that are associated with the aging process and that are comparable to those used in humans. We will test the hypothesis that there is a negative correlation between ability on physical performance measures in healthy older rats and incidence of future ADL-like disability and mortality, and that this decline in physical performance is related to preclinical conditions such as loss of muscle strength. In order to accomplish this we will assess in apparently well functioning older rats: 1) the association of physical performance (swim speed, walking speed, locomotor activity, and muscle strength) with mortality; 2) the onset of age-related declines in ADL-like activities (ambulation and grooming) and their association with mortality; and 3) the association between physical performance measures and overt ADL-like activities in animals. Animals will be tested monthly on a battery of ADL-like and physical performance measures beginning at approximately 24months and continuing for 6 months or until death. The ability of the performance measures to predict future ADL-like disabilities and mortality will be assessed both individually and compositely using a stepwise linear regression model. Future studies will use this model to construct physiological links between aging and ADL-like disability using various models of aging and by studying correlations between age-related changes in biomarkers and functional status. This model will also be used for medical screening of new pharmacological interventions.